Quizmaster

Learning objectives

What you will be able to explain

Which layer controls each cache in the lecture overview?
What does temporal locality mean in the slides?
How is DBMS memory management positioned relative to virtual memory?
Which statement about pages is correct?
In slotted pages, tupleID is represented as:
What is a key drawback of pure column layout (DSM) highlighted in lecture?

Section 01

Which layer controls each cache in the lecture overview? to How is DBMS memory management positioned relative to virtual memory?

Which layer controls each cache in the lecture overview?

The slide distinguishes hardware-controlled CPU caches, OS-controlled file-system cache, and DBMS-controlled buffer cache.

Source: 01_Storage.pdf, page 9

What does temporal locality mean in the slides?

Temporal locality is defined as currently/recently used data likely being reused.

Source: 01_Storage.pdf, page 10

How is DBMS memory management positioned relative to virtual memory?

The slide notes virtual memory behavior and states database systems often have their own memory management, while some rely on OS.

Source: 01_Storage.pdf, page 20

Section 02

Which statement about pages is correct? to What is a key drawback of pure column layout (DSM) highlighted in lecture?

Which statement about pages is correct?

Pages are described as fixed-sized blocks with header and data, and they carry identifiers.

Source: 02_Data_Layouts.pdf, page 10

In slotted pages, tupleID is represented as:

The slotted-page structure explicitly shows tupleID as pageID + slotID.

Source: 02_Data_Layouts.pdf, page 11

What is a key drawback of pure column layout (DSM) highlighted in lecture?

The DSM/column-layout slide states that column layout requires tuple reconstruction.

Source: 02_Data_Layouts.pdf, page 31

Section 03

How is PAX characterized in the slides? to CLOCK is introduced as:

How is PAX characterized in the slides?

PAX is shown as NSM across pages and DSM inside a page.

Source: 02_Data_Layouts.pdf, page 43

Which eviction policies are explicitly listed together?

The eviction-policy slide lists FIFO, LRU, LFU, and CLOCK.

Source: 03_Caching.pdf, page 20

CLOCK is introduced as:

The CLOCK slide describes it as an improvement over FIFO that approximates LRU with lower overhead.

Source: 03_Caching.pdf, page 25

Section 04

In CLOCK eviction search, what happens when reference bit = 1? to A dense index means:

In CLOCK eviction search, what happens when reference bit = 1?

The algorithm says bit=1 gives a second chance: set to 0 and continue; bit=0 is evicted.

Source: 03_Caching.pdf, page 26

What does the prefetching slide emphasize?

The slide notes prefetching future-needed blocks and that sequential asynchronous reads often cost little while saving time.

Source: 03_Caching.pdf, page 33

A dense index means:

Dense indexes are defined as one index entry per tuple/record.

Source: 04_Indexing 1.pdf, pages 12 and 14

Section 05

Sparse indexes are most useful when: to Which B-tree property is explicitly stated?

Sparse indexes are most useful when:

The lecture states sparse indexes (often one entry per block) are useful when records are sorted.

Source: 04_Indexing 1.pdf, page 16

What is true for secondary indexes in this lecture?

The secondary-index slide states it does not define record order and can only be dense.

Source: 04_Indexing 1.pdf, page 23

Which B-tree property is explicitly stated?

B-trees are presented as always balanced, with equal root-to-leaf path length.

Source: 05_Indexing 2.pdf, page 12

Section 06

During insertion, when root split occurs in a B-tree: to In the hash-table model, h(K) returns:

During insertion, when root split occurs in a B-tree:

The insertion algorithm notes that splitting root creates a new root and increases height by one.

Source: 05_Indexing 2.pdf, page 29

For larger data sets, bulk loading into B-trees is improved by:

The bulk-loading slide recommends sorting key-pointer pairs and exploiting order for efficient loading.

Source: 05_Indexing 2.pdf, page 47

In the hash-table model, h(K) returns:

Hash output is defined as an integer from 0 to B-1.

Source: 06_Indexing 3_hash_tables.pdf, page 8

Section 07

Which dynamic hash-table variants are explicitly covered? to Which index families are shown in the multidimensional-index overview?

Which dynamic hash-table variants are explicitly covered?

The dynamic-hash section lists extensible hashing and linear hashing.

Source: 06_Indexing 3_hash_tables.pdf, page 20

Compared to extensible hashing, linear hashing is noted as using:

The linear-hash slide explicitly notes no bucket directory compared to extensible hashing.

Source: 06_Indexing 3_hash_tables.pdf, page 31

Which index families are shown in the multidimensional-index overview?

The overview slide lists these multidimensional index families.

Source: 07_1_Multidimensional Indexing.pdf, page 63

Section 08

For partial-match queries over bitmap indexes, the lecture suggests: to How are attributes handled across levels in kd-trees?

For partial-match queries over bitmap indexes, the lecture suggests:

The bitmap-query slide describes partial-match queries via AND on bitmaps before fetching matching blocks/rows.

Source: 07_1_Multidimensional Indexing.pdf, page 94

Grid-file construction starts by:

The grid-file idea slide states partitioning every dimension into stripes and overlaying multidimensional space with a grid.

Source: 07_2_multidimensional indexing_pt2.pdf, page 2

How are attributes handled across levels in kd-trees?

The kd-tree slide notes that splitting attributes alternate through levels.

Source: 07_2_multidimensional indexing_pt2.pdf, page 20

Section 09

What relationship between logical and physical plans is shown? to Morsel-driven parallelism assigns work by:

What relationship between logical and physical plans is shown?

The lecture shows logical operators with alternative physical implementations (for example hash join vs sort-merge join).

Source: 08_Query Execution.pdf, page 5

Which three processing-model dimensions are explicitly named?

The processing-model slide names these three dimensions explicitly.

Source: 08_Query Execution.pdf, page 10

Morsel-driven parallelism assigns work by:

The slide describes splitting work into morsels and dynamically dispatching them to threads through a task queue.

Source: 08_Query Execution.pdf, page 62

Section 10

Which task belongs to the parser stage? to In cost estimation, output cardinality is modeled as:

Which task belongs to the parser stage?

The parser slide assigns parse-tree construction and syntax correctness checking to parser stage.

Source: 09_Query_Optimization.pdf, page 8

Which action is explicitly listed under preprocessing?

Preprocessing is shown to include semantic checks like relation/view existence, attribute existence, and type correctness.

Source: 09_Query_Optimization.pdf, page 11

In cost estimation, output cardinality is modeled as:

The slide defines output cardinality as input cardinality times selectivity (filter factor).

Source: 09_Query_Optimization.pdf, page 26

Section 11

What is the Selinger-style extension over plain dynamic programming? to A schedule is conflict-serializable iff:

What is the Selinger-style extension over plain dynamic programming?

Selinger-style optimization is described as keeping interesting-order variants, not only one cheapest plan.

Source: 09_Query_Optimization.pdf, page 49

Serializable schedules allow interleaving as long as:

The lecture defines serializable by equivalence to a serial schedule result, not by forbidding interleaving.

Source: 10_Concurrency Control.pdf, page 24

A schedule is conflict-serializable iff:

Conflict-serializable is stated as equivalent to a cycle-free precedence graph.

Source: 10_Concurrency Control.pdf, page 36

Section 12

What does 2PL guarantee, and what can still happen? to What does the STEAL / NO-STEAL policy control?

What does 2PL guarantee, and what can still happen?

The 2PL slide states serializability guarantee while noting deadlocks are still possible.

Source: 10_Concurrency Control.pdf, page 47

Which statement about READ COMMITTED is correct in the slide set?

The isolation-level slide states READ COMMITTED disallows dirty reads but allows fuzzy reads and phantoms.

Source: 10_Concurrency Control.pdf, page 70

What does the STEAL / NO-STEAL policy control?

STEAL policy concerns writing dirty pages from uncommitted transactions; NO-STEAL forbids it.

Source: 11_Recovery.pdf, pages 7 and 8

Section 13

What is a core WAL requirement? to How is speedup defined in parallel DB lecture?

What is a core WAL requirement?

Write-ahead logging requires forcing update log records before writing affected data pages.

Source: 11_Recovery.pdf, page 16

What is the order of phases in the three-pass recovery process?

The recovery process is shown as Analysis pass, then Redo pass, then Undo pass.

Source: 11_Recovery.pdf, page 24

How is speedup defined in parallel DB lecture?

Speedup is defined as sequential runtime divided by runtime of the parallel program on p nodes.

Source: 12_Parallel_DBs_and_Big_data.pdf, page 7

Section 14

Which pairing of parallelism focus is given? to How are replication and fragmentation related?

Which pairing of parallelism focus is given?

The lecture frames inter-query parallelism as key for transactional scenarios and intra-query for analytical scenarios.

Source: 12_Parallel_DBs_and_Big_data.pdf, page 24

How are replication and fragmentation related?

The slide defines replication and fragmentation and explicitly says both can be combined.

Source: 12_Parallel_DBs_and_Big_data.pdf, page 31

Knowledge check

Turn understanding into recall.

The quiz now follows the same concepts in scored form. You can return to this lesson from the quiz whenever a gap appears.

Start the quiz →Back to course